Longitudinal changes in psychosocial constructs and physical activity among adults with physical disabilities
Abstract
Background
Given the importance of physical activity (PA) and the low activity levels among adults with physical disabilities, it is important to understand how temporal changes in psychosocial constructs affect PA changes over time.
Objective/Hypothesis
Examine if changes in the transtheoretical model (TTM) constructs affected changes in PA levels over time.
Methods
One hundred thirty-two adults with physical disabilities, such as multiple sclerosis and spinal cord injuries, completed a web-based survey once every 4 months, for a total of 3 time points, to assess the TTM constructs and PA. Six latent growth curve analyses were conducted using Mplus5 to examine if longitudinal changes in the TTM constructs affected temporal changes in PA levels.
Results
All six hypothesized models fit the sample data well (e.g., χ2 = NS; RMSEA = <.001-.06). In a descending order of significance, the best predictors of the initial levels of PA were the stages of change, the behavioral processes of change, the cognitive processes of change, self-efficacy, and perceived pros. The meaningful predictors of PA changes over time were the initial levels and the slopes of the cognitive processes of change, perceived pros, and the behavioral processes of change. Although the slopes of the stages of change and perceived cons did not have a statistically significant effect on PA changes, their effects approached a medium size (.33 and .38, respectively).
Conclusions
In order to reassure the maintenance of an exercise program, interventionists need to first emphasize cognitive, motivational strategies (cognitive processes of change), such as the importance of PA and positive thoughts about exercise participation as well as exercise benefits (pros) before they implement behavioral strategies (behavioral processes of change), such as social support, goal setting, and self-rewarding.
Keywords: Longitudinal study, Transtheoretical model, Physical activity, Latent growth curve analysis, Physical disabilities
The authors have no conflicts of interest to declare. The authors would like to thank the funding agency of the project, the National Multiple Sclerosis Society (Grant PP1463). Additionally, we would like to acknowledge Dr. Alan Acock, Distinguished Professor of Human Development and Family Sciences at Oregon State University, for his assistance with identifying resources about latent growth curve analyses.
PII: S1936-6574(11)00073-2
doi:10.1016/j.dhjo.2011.09.002
© 2012 Elsevier Inc. All rights reserved.
